Efficient Selection of Biomineralizing DNA Aptamers Using Deep

Dec 16, 2013 - School of Chemistry, University of Leeds, Leeds, United Kingdom LS2 9JT ... Advances in high-throughput DNA sequencing expand the power...
0 downloads 0 Views 5MB Size
Lukmaan A. Bawazer,†,§,^ Aaron M. Newman,†,§,^ Qian Gu,† Abdullah Ibish,†,^ Mary Arcila,† James B. Cooper,† Fiona C. Meldrum,‡ and Daniel E. Morse†,*

ARTICLE

Efficient Selection of Biomineralizing DNA Aptamers Using Deep Sequencing and Population Clustering †

Department of Molecular, Cellular and Developmental Biology, Institute for Collaborative Biotechnologies, and Biomolecular Science and Engineering Program, University of California, Santa Barbara, California 93106, United States, and ‡School of Chemistry, University of Leeds, Leeds, United Kingdom LS2 9JT. §These authors contributed equally to this work. ^Present address: L.A.B, University of Leeds, Leeds, UK, LS1 9JT; A.M.N, Institute for Stem Cell Biology and Regenerative Medicine, Stanford University, Stanford, CA, 94305; A.I., Georgetown University School of Medicine, Washington, DC 20057.

ABSTRACT DNA-based information systems drive the combinatorial optimization processes of natural

evolution, including the evolution of biominerals. Advances in high-throughput DNA sequencing expand the power of DNA as a potential information platform for combinatorial engineering, but many applications remain to be developed due in part to the challenge of handling large amounts of sequence data. Here we employ high-throughput sequencing and a recently developed clustering method (AutoSOME) to identify single-stranded DNA sequence families that bind specifically to ZnO semiconductor mineral surfaces. These sequences were enriched from a diverse DNA library after a single round of screening, whereas previous screening approaches typically require 515 rounds of enrichment for effective sequence identification. The consensus sequence of the largest cluster was poly d(T)30. This consensus sequence exhibited clear aptamer behavior and was shown to promote the synthesis of crystalline ZnO from aqueous solution at near-neutral pH. This activity is significant, as the crystalline form of this wide-bandgap semiconductor is not typically amenable to solution synthesis in this pH range. High-resolution TEM revealed that this DNA synthesis route yields ZnO nanoparticles with an amorphouscrystalline coreshell structure, suggesting that the mechanism of mineralization involves nanoscale coacervation around the DNA template. We thus demonstrate that our new method, termed Single round Enrichment of Ligands by deep Sequencing (SEL-Seq), can facilitate biomimetic synthesis of technological nanomaterials by accelerating combinatorial selection of biomolecularmineral interactions. Moreover, by enabling direct characterization of sequence family demographics, we anticipate that SEL-Seq will enhance aptamer discovery in applications employing additional rounds of screening. KEYWORDS: SELEX . massively parallel sequencing . next-generation sequencing . DNA aptamers . zinc oxide . biomimetic mineralization . AutoSOME . population clustering

D

NA-based high-throughput screening and in vitro evolution encompass an array of biotechnological strategies that mimic molecular aspects of natural evolution. Recently demonstrated as powerful new tools for materials engineering,1,2 these strategies generate individual DNA sequences,3 RNA sequences,4 short polypeptides,5,6 or full proteins (such as enzymes)7 with novel activities. An example of such activity is aptamer behavior, in which a short biopolymer (DNA, RNA, or polypeptide) exhibits tight binding affinity to a molecular or mineral target. The method for producing nucleic acid aptamers is generally known as SELEX, for SElection of Ligands by EXponential enrichment.3,4 SELEX is typically utilized to screen single-stranded (ss) BAWAZER ET AL.

DNA libraries for individual ssDNA aptamer sequences. Toward establishing efficient screening protocols, ssDNA aptamer discovery via SELEX is experimentally advantageous relative to other approaches such as phage-display or cell surface-display polypeptide screening or RNA library screening.2,3 This is because ssDNA aptamers simultaneously exhibit both target binding and information encoding functionality, allowing both genotype and phenotype to be expressed from a single molecule. As a result, bound ssDNA aptamer sequences can be directly enzymatically amplified (after elution from the target), obviating experimental steps needed to link genotype and phenotype that are required in other screening approaches.2 All in vitro evolution strategies, VOL. 8



NO. 1

* Address correspondence to [email protected]. Received for review August 26, 2013 and accepted December 16, 2013. Published online December 16, 2013 10.1021/nn404448s C 2013 American Chemical Society



387–395



2014

387 www.acsnano.org

BAWAZER ET AL.

and/or number,19 limiting their utility for broad demographic analyses. In this study, we used AutoSOME,13 a novel strategy for revealing natural cluster structure in large data sets. As was recently demonstrated,20 AutoSOME can identify representative sequence populations without the need for seed sequences, and without prior knowledge of motif length or cluster number. Thus, the approach described here should both accelerate combinatorial biomolecular discovery for materials engineering applications, and be generally applicable to multiple variations of in vitro evolution.

ARTICLE

including SELEX, rely on the handling of large, diverse libraries of DNA molecules. The molecular population demographics of these libraries change over successive rounds of screening or selection, as families of high activity-encoding DNA sequences become increasingly enriched and prevalent in the library through each successive round of selection. Obtaining sequence information from sampled DNA subpopulations of the screened library is critical for identifying specific subsets of active biomolecules. Until a few years ago, the bottleneck of traditional DNA sequencing methods limited such sampling analyses to a small number of DNA molecules (often